Robert Rector is a senior research fellow in the Heritage Foundation’s domestic policy studies department.

To hear Bill Clinton tell it, there’s no truth to the charges that President Obama gutted welfare reform. The White House, fact-checkers and some journalists have said the same, playing down Obama’s decision to exempt states from the law’s work requirements.

Working closely with members of Congress, I helped draft the work requirements in the 1996 law, and I raised the alarm on July 12, when the Obama administration issued a bureaucratic order allowing states to waive those requirements. The law has indeed been gutted. Here’s how:

The 1996 welfare reform law required that a portion of the able-bodied adults in the Temporary Assistance for Needy Families (TANF) program — the successor to the Aid to Families with Dependent Children program — work or prepare for work. Those work requirements were the heart of the reform’s success: Welfare rolls dropped by half, and the poverty rate for black children reached its lowest level in history in the years following.

But the Obama administration has jettisoned the law’s work requirements, asserting that, in the future, no state will be required to follow them. In place of the legislated work requirements, the administration has stated, it will unilaterally design its own “work” systems without congressional involvement or consent. Any state will be free to follow the new Obama requirements “in lieu of” the written statute.

The administration has provided no historical evidence showing that Congress intended to grant the Department of Health and Human Services (HHS) or any part of the executive branch the authority to waive the TANF work requirements. The historical record is clear and states the opposite; as the summary of the reform prepared by Congress shortly after enactment plainly says: “Waivers granted after the date of enactment may not override provisions of the TANF law that concern mandatory work requirements.”

The members of Congress closely involved in drafting this law have asserted that Obama’s action contradicts the letter and intent of the statute. For 15 years after welfare reform was enacted, no waivers of work requirements were issued by HHS. No such waivers were discussed because it was clear to all that Congress had never provided the department with such waiver authority.

What is it that the administration’s July guidance suddenly seeks to change? At the core of the 1996 law are “participation rate requirements” that ensure that 30 to 40 percent of able-bodied TANF recipients must engage in any of 12 different “work activities” for 20 to 30 hours per week. The administration would exempt states from this requirement and encourage them to operate under alternative performance measures. For example, HHS Secretary Kathleen Sebelius has said that to bypass federal workfare requirements, a state would have to “move at least 20 percent more people from welfare to work compared to the state’s past performance.”

At first blush, a 20 percent increase in “employment exits” sounds impressive. But what does it mean? In the typical state, about 1.5 percent of the TANF caseload leaves the rolls each month because of employment. To be exempt from the federal work requirement, a state would have to raise that number to about 1.8 percent of caseload. This is a minuscule change; as the economy improves, this small increase will occur automatically in most states. Moreover, states keep imperfect employment records of those leaving TANF; many states could easily achieve the required increase through modest improvements in recordkeeping alone.

But here’s the kicker. States have kept statistics on employment exits for decades, and they have always been meaningless as a measure of success. Welfare caseloads always have routine turnover; the larger the caseload, the greater the number of exits, simply because there are more people in the system. Historically, the number of employment exits rises as the caseload rises and falls as the caseload falls. The count of employment exits is at best pointless; at worst, it is a reverse indicator of limiting welfare dependence.

For example, according to the metric of employment exits, the Aid to Families with Dependent Children system was a whopping success: Caseloads soared and the number of employment exits nearly doubled. By contrast, the post-reform TANF program has been a failure, because caseloads fell and employment exits declined. This is why, when the 1996 reform was drafted, the count of employment exits was deliberately excluded as a success measure. It is inherently misleading.

The Obama administration is waiving the federal requirement that ensures a portion of able-bodied TANF recipients must engage in work activities. It is replacing that requirement with a standard that shows that the pre-reform welfare program was successful and the post-reform program a failure. If that is not gutting welfare reform, it is difficult to imagine what would be.